CRAN Release 0.1.0 - Githubissues

ncss-tech / SoilTaxonomy

A System of Soil Classification for Making and Interpreting Soil Surveys

https://ncss-tech.github.io/SoilTaxonomy/

GNU General Public License v3.0

14 stars 2 forks source link

CRAN Release 0.1.0 #16

Closed brownag closed 3 years ago

brownag commented 3 years ago

I want to submit this package for a first CRAN release ASAP.

[x] Run test code coverage analysis and tackle tests for any low-hanging fruit that are untested
[x] Review all existing tests for duplication and thoroughness
[x] explainST need to look for formative element dictionary edge cases -- RE: #15, #5.

[x] Submit to Winbuilder, fix anything
[x] Get into CRAN first submission queue

dylanbeaudette commented 3 years ago

Thanks. Do we need to fully solve #15/#5 before a CRAN submission? I'd like to make sure that most cases are resolved, but a general solution is likely going to take some time / planning.

brownag commented 3 years ago

No, I don't think we need to "solve" them completely. But I personally have spent very little time with the explainST code aside from my verification that it runs on all subgroups.

I essentially want to look closer at the ones that I know are problematic, and figure out why. I am not sure about the extent of the problems at this point... my feeling is it is actually not that much work.

Sure refining definitions could go on forever, but getting known missing entries in the table and ensuring that all multi-word subgroups at least populate with "?" for missing entries shouldn't be too hard, right?

dylanbeaudette commented 3 years ago

No, I don't think we need to "solve" them completely. But I personally have spent very little time with the explainST code aside from my verification that it runs on all subgroups.

I essentially want to look closer at the ones that I know are problematic, and figure out why. I am not sure about the extent of the problems at this point... my feeling is it is actually not that much work.

Sure refining definitions could go on forever, but getting known missing entries in the table and ensuring that all multi-word subgroups at least populate with "?" for missing entries shouldn't be too hard, right?

Works for me. Maybe we can divide / conquer.

brownag commented 3 years ago

We are in pretty good shape now! With a couple recent additions to explainST tests, we are now at 95% coverage package-wide.

brownag commented 3 years ago

We are on CRAN as of 2021-02-25